Short-term leprosy forecasting from an expert opinion survey

Area of Science:

• Public Health
• Epidemiology
• Infectious Diseases

Background:

• Leprosy (Hansen's Disease) remains a significant global health challenge.
• Forecasting disease incidence is crucial for effective public health interventions.
• Expert opinion is a valuable, yet often underutilized, resource in epidemiological forecasting.

Purpose of the Study:

• To assess the accuracy of expert forecasts for leprosy case numbers.
• To compare expert forecasts with statistical modeling methods.
• To explore the relationship between experts' goal-setting beliefs and forecast accuracy.

Main Methods:

• An expert survey was conducted in February 2016, involving 103 respondents.
• Experts provided forecasts for global, national (top 3 countries), and sub-national (Indian states) leprosy cases.
• Forecasts included lower and upper confidence bounds; comparisons were made with regression and exponential smoothing.

Main Results:

• No statistical forecasting method demonstrated superiority over expert opinion.
• Experts' confidence in achieving global interruption and disability reduction goals correlated with forecast accuracy.
• This correlation varied by country, with higher error scores for India and Indonesia, but lower for Brazil.

Conclusions:

• Expert forecasting for leprosy is a practical and simple method for public health.
• Forecasting exercises can be integrated into routine public health surveillance.
• Understanding expert beliefs is important for interpreting forecast accuracy in disease control efforts.